The application of ocean front metrics for understanding habitat selection by marine predators

Abstract

Marine predators such as seabirds, cetaceans, turtles, pinnipeds, sharks and large teleost fish are essential components of healthy, biologically diverse marine ecosystems. However, intense anthropogenic pressure on the global ocean is causing rapid and widespread change, and many predator populations are in decline. Conservation solutions are urgently required, yet only recently have we begun to comprehend how these animals interact with the vast and dynamic oceans that they inhabit. A better understanding of the mechanisms that underlie habitat selection at sea is critical to our knowledge of marine ecosystem functioning, and to ecologically-sensitive marine spatial planning. The collection of studies presented in this thesis aims to elucidate the influence of biophysical coupling at oceanographic fronts – physical interfaces at the transitions between water masses – on habitat selection by marine predators. High-resolution composite front mapping via Earth Observation remote sensing is used to provide oceanographic context to several biologging datasets describing the movements and behaviours of animals at sea. A series of species-habitat models reveal the influence of mesoscale (10s to 100s of kilometres) thermal and chlorophyll-a fronts on habitat selection by taxonomically diverse species inhabiting contrasting ocean regions; northern gannets (Morus bassanus; Celtic Sea), basking sharks (Cetorhinus maximus; north-east Atlantic), loggerhead turtles (Caretta caretta; Canary Current), and grey-headed albatrosses (Thalassarche chrysostoma; Southern Ocean). Original aspects of this work include an exploration of quantitative approaches to understanding habitat selection using remotely-sensed front metrics; and explicit investigation of how the biophysical properties of fronts and species-specific foraging ecology interact to influence associations. Main findings indicate that front metrics, particularly seasonal indices, are useful predictors of habitat preference across taxa. Moreover, frontal persistence and spatiotemporal predictability appear to mediate the use of front-associated foraging habitats, both in shelf seas and in the open oceans. These findings have implications for marine spatial planning and the design of protected area networks, and may prove useful in the development of tools supporting spatially dynamic ocean management.